1) Field of the Invention
The present invention relates to a programmable information processing apparatus (e.g. logic modules, integrated circuits, units, information processors, and computing systems), and more particularly to an information processing apparatus with a self-repair function.
2) Description of the Related Art
Recently, there has been a demand to increase the reliability of information processing systems. In response, highly-reliable information processing systems that can recover, by themselves, after a trouble occurrence, without depending on an operator, have been proposed.
Some such information processing apparatuses have a self-repair function, by which, even if trouble occurs in hardware, the system can be recovered by separating the fault portion or operating the substitutive circuit, without stopping the operation of the entire system and asking for assistance.
A system that uses multiplexing building blocks (forming elements) and selects the backup system when a trouble occurs in the main system, is shown in FIG. 31.
The information processing system with self-repair function, shown in FIG. 31, is a typical highly-reliable information processing apparatus forming, for example, a non-stop computer. Referring to FIG. 31, numeral 201 represents a CPU board, 202 represents a LAN board, and 203 represents a line board. These blocks, each formed on a printed board, are multiplexed by plural printed board units.
The printed board units, each formed of plural integrated circuits, function respectively as the CPU board 201, the LAN board 202, and the line board 203.
Numeral 204 represents a back plane board which interconnects the CPU boards 201, the LAN boards 202, and the line boards 203. The numeral 205 represents a LAN transceiver, 206 represents a line switcher, and 207 represents a network connected to the line switcher 206.
In the information processing apparatus with self-repair function shown in FIG. 31, when a specific printed board unit, among the multiplexed printed board units, malfunctions, the operation of the printed board unit is stopped to operate the spare printed board unit, whereby the entire system continues its operation.
When a line board 203 experiences trouble during operation, it is stopped and the line switch 206 operates to select a spare line board 203, thus continuing the operation of the entire system. However, even if a single integrated circuit, acting as a component (forming element) in the system, is broken, the entire system must be replaced, requiring a servicing system and a backup system The entire system is large in size and high priced.
Where the system operation is switched from the servicing system to the backup system, because of trouble or fault of the servicing system, the servicing system must be manually repaired or parts must be replaced during the successive operation of the backup system. Hence the first system cannot function when the backup system is troubled during the repair of the servicing system (multiple troubles). This self-repair-type information processing system does not require the replacement of the entire system in the case of an integrated circuit broken. However, each printed board unit must be entirely multiplexed. Hence requiring a large sized system.
Another type of system is one in which a fault portion is replaced with a spare member in small unit, as typified by substantive memories arranged in memory circuits or substantive sector regions prepared in magnetic disk units. This system has an advantage in that the spare member can be miniaturized. However, since the spare member is a specific portion having the memory function of a memory or magnetic disk, other portions with no memory functions troubled cannot be replaced.
Particularly, as to a system used within an artificial satellite orbiting in the outer space, it is almost impossible to replace and repair externally fault parts. It is desirable to facilitate repair internally by the system itself as much as possible. However, the first and second systems have a limited ability to self-repair. A local fault results in a complete replacement of each system or unit, even if the remaining components are alive. Hence many normal components in the system cannot be utilized at all.